| 02:50 | <GPHemsley> | GPHemsley: you have write access |
| 07:26 | <hsivonen> | Do I understand correctly that there are no normative requirements on the user agent style sheet(s) for media documents? |
| 07:27 | <Ms2ger 💉💉> | I don't recall seeing any when I last looked, but that was probably five years ago at least |
| 07:30 | <hsivonen> | Thanks. |
| 07:48 | <annevk> | smaug: unless you have an objection to the things presented in https://github.com/whatwg/dom/pull/1009 I'm inclined to merge it as it meets all the criteria as far as I can tell |
| 07:49 | <annevk> | GPHemsley: ah, yeah, we have main branch protection, let me add you to the exception list (it's currently empty) |
| 07:51 | <annevk> | GPHemsley: status checks are enforced at the moment so any changes will have to go through a PR; reviews are not |
| 07:53 | <annevk> | hsivonen: I sometimes wonder if we should make normative requirements on them, at least when they are not top-level or appear in a controlled popup |
| 08:39 | <smaug> | annevk: added a comment. I think that is fine addition to DOM, but the performance tests look rather limited to me, and they seem to mostly test some inefficiency in some particular Range implementation. |
| 09:21 | <zcorpan> | hsivonen: are you able to review https://github.com/whatwg/html/pull/5959 ? |
| 09:21 | <hsivonen> | hsivonen: are you able to review https://github.com/whatwg/html/pull/5959 ? |
| 09:42 | <hsivonen> | Looking |
| 09:45 | <zcorpan> | hsivonen: thanks. speculative parsing occurring when there isn't a blocking <script> is new to me |
| 09:46 | <hsivonen> | Hmm. I thought that Firefox can start speculative fetching the resources found by non-speculative parsing before the corresponding nodes end up in the DOM, but now that you say that, I need to check if that actually happens. |
| 09:53 | <hsivonen> | Hmm. I thought that Firefox can start speculative fetching the resources found by non-speculative parsing before the corresponding nodes end up in the DOM, but now that you say that, I need to check if that actually happens. |
| 09:54 | <hsivonen> | Not sure how you'd observe this from a script, but this should happen if you have a huge scriptless document with a bunch of images. |
| 10:20 | <annevk> | smaug: yeah, I was wondering how specific it would be to WebKit/Chromium |
| 10:33 | <sujaldev> | sujaldev: Parsing is definitely hard at first! I've been deep in parsing for a long time (I wrote the CSS Syntax spec), so my difficulty judgement is a bit skewed. ^_^ Which should be an indicator of how hard layout is, tho. |
| 11:14 | <freddy> | I liked the suggestion (somewhere way above) to find out more about non-conforming layouts (e.g., from browsers like lynx) |
| 11:39 | <sideshowbarker> | https://github.com/mdn/content/issues/8486 could use help from somebody with good WebGL familiarity. Basically adding a few more sentences of explanatory text to an existing tutorial. |
| 11:47 | <zcorpan> | hsivonen: ok, yeah, speculative fetches from non-speculative parsing isn't news and the spec should allow for that to happen. I thought you asked for speculative parsing to be allowed without blocking <script>s |
| 11:49 | <zcorpan> | hsivonen: as for speculative fetches while looking for a meta charset... if we are to change the spec to run the tree builder while looking for meta charset, that would be more natural to spec speculative fetches |
| 11:51 | <sideshowbarker> | https://github.com/mdn/content/issues/8448 could use some help from somebody with good 2dcontext familiarity. Basically just needs a comment to confirm whether or not what the OP is asserting is actually correct. |
| 12:22 | <hsivonen> | hsivonen: as for speculative fetches while looking for a meta charset... if we are to change the spec to run the tree builder while looking for meta charset, that would be more natural to spec speculative fetches |
| 12:39 | <zcorpan> | ty hsivonen ! that green tick feels magical, 1 year in 😀 |
| 12:39 | <hsivonen> | ty hsivonen ! that green tick feels magical, 1 year in 😀 |
| 12:41 | <zcorpan> | no worries, it's slowness on my end as well I guess. maybe we can try fast-track changes in the future and sync better |
| 13:08 | <sujaldev> | I liked the suggestion (somewhere way above) to find out more about non-conforming layouts (e.g., from browsers like lynx) |
| 13:11 | <sideshowbarker> | https://stackoverflow.com/questions/68993824/cors-issues-with-window-open-and-oauth2 |
| 17:15 | <Domenic> | annevk: want to review https://github.com/whatwg/html/pull/7000 ? |
| 17:24 | <bakkot> | Domenic: re: https://github.com/tc39/proposal-iterator-helpers/issues/122#issuecomment-909427809: I think it's actually stream[Symbol.asyncIterator]().map(...), which suggests maybe streams (or all web-platform iterables?) ought to have a more conveniently named method which exposes the iterator |
| 17:24 | <bakkot> | wait, unless there is such a method and I don't know about it |
| 17:25 | <Domenic> | bakkot: ah yeah, you're right. The method is .values() |
| 17:25 | <bakkot> | ah, excellent |